# News this Week

Science  30 Jul 2010:
Vol. 329, Issue 5991, pp. 496
1. Archaeology

# Copper Mine Threatens Ancient Monastery in Afghanistan

1. Andrew Lawler

VIENNA—A decade after the Taliban destroyed the famous Bamiyan Buddhas—two massive statues that have stood sentry in an Afghan valley for 1500 years—archaeologists are warning that Afghan antiquities are again at risk. This time the threat comes from a venture blessed by the Western-backed Afghan government. A Chinese company intends to blow up an ancient Buddhist monastery south of Kabul to make way for a massive copper mine. The plan has sparked outrage among Afghan and French archaeologists, who have recently uncovered more than 100 statues within a large religious complex that includes seven stupas, or tombs built to house the relics of saints.

In their first public talk on the finds at a recent meeting here,* excavation leaders called for an international meeting to galvanize support for the site. And French officials intended to raise concerns about the impending demolition at a meeting of foreign ministers from around the world in Kabul last week. The controversy pits Afghanistan's desperate need for revenue against its ancient heritage as an Asian crossroads. “This will have to be decided by [Afghan President Hamid] Karzai,” says Philippe Marquis, head of the French archaeological mission to Afghanistan.

Marquis says plans to dynamite the monastery in April were postponed but have not been canceled. The United States and its allies are pushing for increased foreign investment in Afghanistan, and a 14 June report issued by the U.S. Defense Department estimated that $1 trillion of minerals such as copper, iron, cobalt, and lithium lie under eastern Afghanistan alone. Some geologists think Mes Aynak holds the world's second largest deposit of copper, and the$3 billion contract with Beijing's China Metallurgical Group Corp. to mine it is the most lucrative in Afghanistan's history.

Located in a mountainous region 40 kilometers southeast of the capital, Mes Aynak is a hill topped by a 4500-square-meter monastery. Although the site was spotted by archaeologists in the 1960s, it was never excavated. During the late 1990s, the hill was home to an al-Qaida training camp, according to the 2004 report by the National Commission on Terrorist Attacks Upon the United States. In recent years, looters have damaged much of the monastery complex in the search for antiquities, says Nader Rassouli, director of Afghanistan's National Institute of Archaeology in Kabul, which is also participating in the current excavations.

“The site is huge, and we have amazing remains,” he said at the meeting. The monastery flourished from as early as the 2nd century B.C.E. until at least the 6th century C.E., although it may have continued as a settlement until as late as the 9th century C.E., says Marquis. The joint Afghan-French team began salvage work last summer, halting in November due to the severe winter at the 2400-meter altitude. Among the finds are three dozen clay statues, including a reclining Buddha 5 meters long, as well as dozens of wooden and stone Buddhas. Rassouli estimates that the site covers 100 to 400 hectares, including two areas yet to be excavated. As the archaeologists work, Chinese engineers are busy building a railroad, housing, and a power plant nearby, in preparation for mining, according to Afghan government statements on the project.

Two millennia ago, this region served as a critical conduit in the spread of Buddhism to Central Asia and China, says T. Richard Blurton, an archaeologist and curator at London's British Museum who has excavated in Afghanistan. He says Mes Aynak could provide new data on both the origin and demise of the religion here. Researchers now believe that as late as the 7th century C.E., when Islam arrived in the area, Buddhism was still making inroads as far west as Iran and as far north as Turkmenistan. “It's quite tantalizing to consider how Buddhism coexisted with the new religion,” Blurton says. There are also Hindu deities from that late period at Tepe Sardar, a large monastery located in the eastern Afghan city of Ghazni. Further evidence from Mes Aynak could help provide a new picture of religious blending at an important historical juncture, says Blurton.

The ancient monks of Mes Aynak apparently knew they were sitting on a large copper deposit, because there are signs of mining throughout the monastery's long history, says Marquis. Soviet geological surveys in the 1970s rediscovered the mineral deposit, but the outbreak of war in 1979 prevented its development. In May 2008, the state-owned China Metallurgical Group signed a contract to begin extracting copper. “This project provides high levels of income for the government and opportunities for more than 6000 people and for thousands of others indirectly,” said Ibrahim Adel, the Afghan minister of mines at the time of the signing ceremony. The Washington Post reported in November that Adel was suspected of taking a $30 million bribe from the Chinese company. Adel denied the allegation but resigned in January; no charges have been filed. Geologists estimate that there are 13 million tons of ore at Mes Aynak. A feasibility study will not be completed until early next year, but Afghan officials estimate that the Chinese will extract 200,000 tons of copper and provide Afghanistan with up to$400 million in annual revenues. Requests for comment about the monastery demolition from both the Afghan Ministry of Mines and the Chinese company went unanswered. Marquis compares the plan to destroy Mes Aynak with the Taliban's infamous 2001 destruction of the Bamiyan Buddhas. “Karzai is the only one who can say no,” he says. In Marquis's view, copper mining and the monastery can coexist by creating a protected archaeological area that eventually could generate tourism income. That compromise may prove a tough sell in a country under intense pressure to produce short-term revenue.

Meanwhile, Rassouli hopes to sponsor an international meeting to drum up support for the ruin. “Time is running short,” he says. “This place is going to be destroyed in a few months, and we need to find another solution—or the site is doomed.”

• * European Association for South Asian Archaeology and Art, 4–9 July 2010.

2. 2011 Budget

# NSF Funding Request Faring Well in Congress

1. Jeffrey Mervis

Last week, a Senate committee declared the U.S. National Science Foundation (NSF) to be “the finest science foundation in the world.” And federal legislators seem prepared to back up their kind words with hard cash. While the 2011 budget requests of other U.S. research agencies are struggling to stay afloat in choppy fiscal seas, so far NSF's is sailing through Congress largely intact.

Last week, the Senate appropriations committee gave NSF all but $71 million of the$498 million increase that it had requested for 2011. A few weeks earlier, a House of Representatives spending panel endorsed the entire 7.2% increase, which would boost NSF's budget next year to $7.4 billion. That's also the target for the first of a 3-year reauthorization of the 2007 America COMPETES Act adopted in late May by the House and approved last week by the Senate Commerce, Science, and Transportation Committee, which threw the above verbal bouquet to the science agency. Both NSF and the Department of Energy's (DOE's) Office of Science are included in the COMPETES reauthorization, which would keep their budgets on a 10-year doubling path as well as give them program guidance. But Senate appropriators have provided the DOE office with only half of a much smaller (4.4%) increase it has requested. What's more, a House spending panel earlier this month told DOE to make do in 2011 with its current budget of$4.9 billion (see graphic).

Both NSF and DOE are still a long way from the finish line. Their spending bills must move through each house and then be reconciled, most likely after the November elections. That timetable may also apply to the reauthorization bills. Although the House version (H.R. 5116) is a top priority for the retiring chair of the House science committee, Representative Bart Gordon (D–TN), the Senate may run out of time before it completes work on its bill (S. 3605).

In the meantime, NSF is riding high in the eyes of key legislators from both parties. An expanding budget would be no small accomplishment at a time when Republicans and many Democrats are demanding that Congress slash federal spending to reduce the $1.4 trillion deficit. “The one thing that might keep us from passing a bill is concern over the level of authorization,” Senator Kay Bailey Hutchison (R–TX), the ranking member of the Senate science panel, told Science after last week's markup of S. 3605. “I think we should authorize basically level funding. But I think that NSF deserves a higher level than the others because of its excellent record of supporting basic research.” Adds committee chair Senator Jay Rockefeller (D–WV), “I like our authorization levels. [That level of support is] vital to our country's future.” To be sure, NSF has taken a few lumps this year. Its clumsy handling of employees found to be using their computers to view and circulate pornography triggered a sharp rebuke in the House version of the COMPETES bill (Science, 21 May, p. 960). And last week, Senate appropriators told the agency to abandon its plan to fold three programs aimed at boosting the number of minority students in science and engineering into one initiative (Science, 23 July, p. 376). “One size will not fit all,” notes the report accompanying the spending bill. Science lobbyists are keeping their fingers crossed that the good vibrations will continue once Congress returns from its August recess. “I am very pleased with the numbers and the accolades for NSF,” says Samuel Rankin of the Coalition for National Science Funding in Washington, D.C. “I will remain cautious, however, until an [appropriations] bill is signed.” 3. Volcanology # Is China's Riskiest Volcano Stirring Or Merely Biding Its Time? 1. Richard Stone CHANGBAI MOUNTAIN, CHINA—On the north slope of Changbai Mountain, throngs of tourists jostle at a precipice to catch a glimpse of Sky Pond, the volcano's huge caldera lake several hundred meters below. For Koreans who revere this mountain on the border of China and North Korea as the birthplace of the Korean nation, no pilgrimage is complete without a peek at Sky Pond. They'd better enjoy the view while it lasts. Last month, a South Korean geologist declared that the 2744-meter-tall volcano, known as Mount Paektu to Koreans and Tianchi caldera to Chinese, is showing signs of waking from a century-long slumber. North Korean scientists, too, are worried about recent geophysical anomalies observed at the mountain, Science has learned. But Chinese volcanologists, who have a sophisticated monitoring system set up at Changbai, discount the threat and insist that the volcano is quiet. South Korea plans to hold a meeting with China and Japan in November to “get to the bottom of the matter,” says an official with the Korean Meteorological Administration (KMA) in Seoul who requested anonymity because of the issue's sensitivity. An eruption of Changbai could have devastating consequences. Of China's 14 active volcanoes, “Changbai poses the biggest hazard and therefore is our most dangerous one,” says Xu Jiandong, director of the China Earthquake Administration's (CEA's) Active Volcano Research Center in Beijing. Part of the risk is the threat of lahars from the huge lake in the 5-kilometer-wide caldera. About 100,000 people live on or near Changbai's slopes, and in summertime the region is packed with tourists. As volcanoes go, Changbai is in the big leagues. In a violent, explosive eruption around 1000 C.E.—dating is not precise, so it's called the “millennium eruption”—the volcano disgorged up to 30 cubic kilometers of magma, or 10 times as much as the famed Krakatoa eruption in 1883. Pyroclastic flows covered an area extending roughly 50 kilometers from Tianchi caldera. In the past 2000 years, only the 1815 eruption of Tambora in Indonesia rivals Changbai, which dumped ash as far away as northern Japan, says Tsuyoshi Miyamoto, a geologist at Tohoku University in Miyagi, Japan, who has studied past eruptions at Changbai. Smaller eruptions have occurred at roughly 100-year intervals since then, with the last one in 1903. At a KMA seminar in Seoul last month, Yun Sung-hyo, a geologist at Pusan National University, citing Chinese data, said signs indicate “that Mount Paektu may have an eruption in the near future,” according to a report by Yonhap News Agency. And in recent months, North Korean geologists who monitor the volcano have sought cooperation with Western volcanologists, including a joint expedition to the volcano, training abroad for North Korean scientists, and upgrades to a research station at Paektu, says a North Korean official who requested anonymity. No cooperation has yet gotten off the ground. Chinese volcanologists keep a close eye on Changbai; they say an eruption does not seem imminent. From 2002 to 2005, the volcano's magma chamber evidently was filling, says Xu. In quiet periods, seismometers register about 100 small tremors a year around the mountain, he says. In 2003, Changbaishan Tianchi Volcano Observatory recorded 1293 tremors. By 2006, the frequency had receded to background rate, and in the first 6 months of 2010 the station recorded 58 tremors. The mountain rose about 6.8 centimeters in 2003 and 2004, and the ratio of helium-3 to helium-4—“a good indication of magma activity because the source is the mantle,” Xu says—nearly doubled during that time. Changbai has been largely quiet since 2005, says Xu, who expects it to kick into another “active phase” again in the next few years. “But that doesn't mean it will erupt,” he says. Miyamoto agrees and notes that it is extremely unlikely that the next eruption would rival the millennium eruption. Korean scientists are wary. One bone of contention is data sharing. “We can only get earthquake data from China, and it's not realtime,” says the KMA official. Data sharing will be a main topic at the tripartite meeting, he says. Xu says that data sharing “is a sensitive question.” China will allow foreigners access to data if they “can ensure CEA that it will be used for scientific purposes only,” he says. Part of the problem is the proximity to the North Korean border. During a visit to Changbaishan Tianchi Volcano Observatory on the north slope of Changbai last week, a staff member declined to answer questions from Science on the grounds that “the observatory's work is a state secret.” CEA has proposed a major upgrade of its Tianchi caldera monitoring network to begin next year. One dark spot is the Korean side of the volcano. A few years ago, China proposed building a seismic station for North Korea and integrating observations, says Xu. The offer was declined. “They even didn't want us to build more observatory stations on our side of the border,” Xu says. In the meantime, Changbai's magma chamber continues to fill. 4. High Energy Physics # Fermilab Physicists Don't See Higgs, Argue They Should Keep Looking 1. Adrian Cho PARIS—This year's International Conference on High Energy Physics was a case study in irony. The meeting was billed as the coming-out party for the Large Hadron Collider (LHC), the gigantic European atom smasher that started taking data in March, but the buzz surrounded results from the older Tevatron collider at Fermi National Accelerator Laboratory (Fermilab) in Batavia, Illinois. Quelling rumors, Fermilab physicists said they had not yet spotted a long-sought particle called the Higgs boson. Nevertheless, their improved nonobservation has them pushing to run the Tevatron three extra years, through 2014—a plan that officials at Fermilab itself do not necessarily favor. Higgs bosons lurking “virtually” in the vacuum are thought to drag on other particles to give them mass. But physicists' standard model of the fundamental particle does not predict how much the Higgs itself will weigh, so scientists must search for it. Physicists know from previous experiments that the Higgs most likely weighs between 114 and 185 giga-electron volts (GeV), or between 121 and 197 times as much as a proton. Last year, Fermilab researchers showed that it likely does not weigh between 162 GeV and 166 GeV. Now, they've widened that “exclusion window” to between 158 GeV and 175 GeV (see figure). That advance suggests that Fermilab researchers have nearly enough data and sensitivity to spot the Higgs if it's anywhere in the 114-to-185-GeV range, says Fermilab's Dmitri Denisov, co-spokesperson for the 500-member team working with the D0 particle detector. “Everywhere, we are very, very close,” he says. If the Tevatron smashed protons into antiprotons through 2014, researchers would collect 65% more data than they will have by the end of next year. That should be enough to glimpse the Higgs—if it's there—before physicists at the LHC can. “There's definitely a window of opportunity here,” Denisov says. That's because the LHC, housed at the European particle physics laboratory CERN near Geneva, Switzerland, will shut down at the end of 2011 to repair thousands of unreliable solder connections between its superconducting magnets. In September 2008, the LHC broke down during preliminary tests when one of those connections melted. After 14 months of repairs, CERN officials decided to run the accelerator at no more than half-energy—still 3.5 times the Tevatron's energy—until all of the suspect connections could be resoldered (Science, 12 February, p. 766). That shutdown will stretch to 15 months to allow workers to make other modifications as well, says Stephen Myers, CERN's director of accelerators and technology. Physicists working with the two biggest particle detectors at the LHC, ATLAS and CMS, say they can overtake the Tevatron experiments with the data they hope to collect by the end of next year. Marat Gataullin, a CMS member from the California Institute of Technology in Pasadena, showed projections that CMS alone would surpass D0 and Fermilab's other particle detector, CDF, in their combined sensitivity to a Higgs weighing more than 135 GeV. But Giampiero Passarino, a theorist at the University of Turin in Italy, says he's skeptical. “By the shutdown? I doubt it.” Running the Tevatron through 2014 would cost about$100 million, says Robert Roser, co-spokesperson for the 600-member team working with CDF. But finding the money would be tough, especially as the lab is building a new $260 million neutrino experiment called NOvA and is developing plans for other projects, says Young-Kee Kim, Fermilab's deputy director: “If we run the Tevatron, then there's a clear impact on NOvA immediately.” Still, Kim says, lab officials are taking the proposal seriously. Some European physicists say they favor running the Tevatron longer. “In an ideal world, I think it would be good to have the Tevatron running in parallel with the LHC,” says Guido Tonelli of the University of Pisa in Italy, spokesperson for the 3170-member CMS team. “For the next 2 or 3 years, the two machines will be complementary.” The price may be too high, some Americans say. “I don't see anybody coming along with a check for the$100 million or whatever it costs to run the Tevatron for 3 years,” says Melvyn Shochet, a physicist at the University of Chicago in Illinois and chair of the U.S. government's High Energy Physics Advisory Panel. “So Fermilab has to consider what it's going to give up” to run the Tevatron. Fermilab's physics advisory committee should weigh in next month.

5. ScienceInsider

# From the Science Policy Blog

Senate Majority Leader Harry Reid (D–NV) declared the effort to pass a mandatory cap on comprehensive greenhouse gas emissions dead as he couldn't muster a bipartisan supermajority of 60 lawmakers. ScienceInsider interviewed a pollster whose data suggested one reason was ignorance: Many Americans didn't even know what cap and trade, the central element of such a limit, was. ScienceInsider included posthumous remarks of climatologist Stephen Schneider on the long effort to pass such a law. And it reviewed a study that said that federal and state rules already on the books—if enforced aggressively—would get the United States close to President Barack Obama's stated goal of cutting emissions by 17% by 2005.

Italian archaeologists were furious after archaeological samples turned blue during storage in a former military facility near Verona. Now a new analysis by mineralogist Gilberto Artioli of the University of Padua in Italy suggests an organic molecule is responsible. He had been asked to analyze the stone tools by archaeologist Laura Longo, the former curator of Verona's Natural History Museum. Longo was suspended without salary recently after she complained about the handling of the artifacts.

Duke University has suspended three cancer clinical trials after allegations that a university scientist on the studies embellished his resume, … while the National Institutes of Health seeks to close a loophole on a conflict-of-interest rule that covers scientists changing jobs.

A group of scientists who study the coasts have lobbied Thad Allen, who heads the federal response to the gulf oil spill, to halt coastal engineering projects intended to protect ecosystems there. Sand berms built in the water to stop oil from striking wetlands have started to erode. A plan to “armor” the berms to halt that problem could be even more disruptive, the scientists say in a letter.

For more science policy news, visit news.sciencemag.org/scienceinsider.

6. International Conference

# Money Woes Cast Shadow Over HIV/AIDS, But Ray of Light in South Africa

1. Jon Cohen

VIENNA—When Nelson Mandela was inaugurated as the president of postapartheid South Africa, he promised that never again would his country “suffer the indignity of being the skunk of the world.” At the 18th International AIDS Conference held here last week, the South African government—which from 1999 to 2008 had a president who questioned whether HIV even caused disease and a health minister who advocated lemons and garlic for treatment—declared that it is no longer the skunk of the AIDS world. “In the past, South Africa has been the subject of much criticism at these conferences for being a highly divided country on its approach to the HIV and AIDS pandemic,” South African Minister of Health Aaron Motsoaledi told the attendees. “Today, we are guided by science.”

“All of us are thrilled by the new direction in South Africa,” said former U.S. President Bill Clinton, echoing the sentiments of the nearly 20,000 participants. Indeed, the news from South Africa was a bright spot in what was otherwise a gloomy discussion of the funding shortfalls that threaten to slow the remarkable progress against HIV/AIDS made during the past 6 years.

The sea change in South Africa—which has an estimated 5.5 million infected people, more than any other country—was especially welcomed because of a growing sentiment that the hardest hit nations must do more to help themselves. As Motsoaledi explained, under the leadership of President Jacob Zuma, last April South Africa launched an ambitious effort to test 15 million people for HIV and expects to start a half-million new people on antiretroviral treatment by June 2011. The HIV/AIDS budget last year increased by 30%, and, he stressed, 83% of the $2.33 billion spent on all related efforts in the country comes from government coffers. Michel Sidibé, head of the Joint United Nations Programme on HIV/AIDS (UNAIDS), urged Zuma last September at the U.N. General Assembly to attack HIV/AIDS aggressively. “I told him the major challenge for any leader is legacy,” Sidibé told Science. “What could be more compelling to your people than saying that during my period, I want to make sure that no babies are born with HIV and no mothers will die from HIV?” But Sidibé didn't expect his lobbying to work, especially because Zuma is married to several women and in 2006 was put on trial (and acquitted) for allegedly raping an HIV-infected woman. Sidibé says he was “amazed” by Zuma's response. South African researchers also unveiled what was widely seen as the meeting's most important scientific finding, which was simultaneously reported online in Science (www.sciencemag.org/cgi/content/abstract/science.1193748). A large clinical study by South African researchers found that a vaginal microbicide for the first time unequivocally protected women from becoming infected by HIV. The gel, which contains the anti-HIV drug tenofovir, also unexpectedly halved the risk of becoming infected by herpes simplex virus-2. HSV-2 infection eases transmission of HIV, so the microbicide packs a one-two punch. (The co-authors said the HSV-2 results were left out of the Science paper for space reasons.) But for most of the weeklong conference, money woes ruled the day. According to UNAIDS, the effort to combat HIV/AIDS in resource-limited countries in 2008 cost$15.6 billion, 70% of which came from international aid. As a result, more than 5 million people now receive lifesaving anti-HIV drugs. But the global financial crisis flattened budgets (see graph) and forced treatment programs to stop enrolling new people, raising concerns about sustaining—let alone expanding—the efforts under way. UNAIDS now calculates that 10 million people urgently need treatment, which would cost an additional $10 billion each year to provide. “The challenge is not finding the money; the challenge is to find the priorities,” said conference chair Julio Montaner of the University of British Columbia in Canada. “When there is a Wall Street emergency or an oil spill, billions upon billions of dollars are quickly mobilized.” Montaner and others say the most immediate way to address the funding shortfall is to increase efficiency by cost-saving tactics such as shifting tasks from doctors to nurses, better integrating HIV/TB care, and starting treatment earlier so people don't require hospitalization. But Kevin De Cock, head of the Center for Global Health at the U.S. Centers for Disease Control and Prevention, warned that efficiencies alone can't solve the problem. “You squeeze and you squeeze and then no more juice comes out,” said De Cock. No one had a concrete plan for finding new funds, but many hope that wealthy countries will contribute more to the main multilateral financing mechanism, the independent Global Fund to Fight AIDS, Tuberculosis and Malaria. Austria, the meeting's host, especially took a drubbing for not being more generous. Some conference goers were deeply disappointed that U.S. President Barack Obama had not done more to address the funding shortfall and worry that he is shifting funds away from HIV/AIDS to maternal and child health in his new$63 billion, 6-year Global Health Initiative. The concern stems in part from a 2008 article in the Journal of the American Medical Association that argues for just such a shift, co-authored by Ezekiel Emanuel, a medical ethicist, brother to Obama chief of staff Rahm Emanuel, and a special adviser to the president's Office of Management and Budget.

“The president didn't write that article,” Emanuel told Science, noting that he penned it before joining the Administration. “And the president submitted budgets that obviously were very different.” Indeed, Obama's budget increases global HIV/AIDS funding for fiscal year 2011 by 3.5%, and a new UNAIDS/Kaiser Family Foundation analysis shows that the United States is the world's largest donor to AIDS efforts, contributing 58% of all disbursements to developing countries in 2009.

Many innovative ideas were bandied about, foremost among them the financial transaction, or Robin Hood, tax that would take a tiny percentage of funds from banks each time they moved large sums to another institution. UNAIDS's Sidibé, an economist by training, and Michel Kazatchkine, who heads the Global Fund, both strongly backed the idea. But Bill Gates expressed doubts. “I know a lot of experts who say it's not workable,” said Gates at a press conference, noting that it would be difficult to enforce. Kazatchkine disagreed, saying it could be organized by currency, so that, say, the euro group did its own policing. “The technical feasibility is there,” he said.

One obvious cost-saving idea that has been much discussed in the backrooms of Washington and Geneva didn't make it onto the meeting agenda: merging the two biggest players, the Global Fund and the U.S. President's Emergency Plan for AIDS Relief (PEPFAR). Kazatchkine and U.S. Ambassador Eric Goosby, who heads PEPFAR, told Science that they think the strategy makes sense. “We're working hard on making that perhaps ultimately happen some day, but happening on some scale as soon as possible,” said Kazatchkine, contending that the structure of the Global Fund is better in the long run than that of PEPFAR because it responds to proposals from affected countries. Goosby says they have already begun to mesh projects in some countries. “People are beginning to take down the walls between these efforts,” said Goosby.

Washington, D.C., will host the meeting in July 2012, which will take these money woes literally to the White House's doorstep.

7. ScienceNOW.org

# From Science's Online Daily News Site

Bacteria Bring Leaves Back From the Dead When the leaves they depend on turn a deathly yellow in the fall, leaf miner moths (Phyllonorycter blancardella) perform CPR. Even as the rest of the leaf wilts, the patch surrounding a leaf miner larva stays a photosynthetically active green. A new study shows that this green island effect is thanks to bacteria within the grubs.

Ecologist David Giron of the Université François Rabelais in Tours, France, and colleagues dosed some female leaf miners with oral antibiotics to kill their internal microbes. Then the insects laid eggs. Larvae from untreated mothers were able to form green islands, but larvae from dosed moms were not, the team reported in the Proceedings of the Royal Society B. The hero could be a microbe of the genus Wolbachia, which carries a gene that, in plants, makes death-delaying hormones called cytokinins.

Earth as an Extrasolar Planet Astronomers have vetted a new method for finding habitable planets—by testing it on Earth. When a planet transits, or passes between its star and Earth, components of the planet's atmosphere subtract some wavelengths from the star's light and add others. By training a spectrograph on this light, scientists can tease out the composition of the planet's atmosphere and look for life-sustaining elements.

To see if this approach could work on a small, Earth-like world, astrophysicist Alfred Vidal-Madjar and colleagues at the Institut d'Astrophysique de Paris decided to try it on a likely candidate: Earth. Taking the moon's-eye-view, the researchers looked at the sunlight filtered through Earth's atmosphere as it passed between the sun and moon during a lunar eclipse. The method was a success: The researchers detected ozone, oxygen, nitrogen, and sodium in the reflected light from Earth's atmosphere, they report in a paper accepted for publication in Astronomy & Astrophysics.

'Locked-In' Patients Can Follow Their Noses Patients who can't move, speak, or even control their blinking—known as “locked-in”—could communicate with the outside world through sniffing, thanks to a new technology.

Neurobiologist Noam Sobel of the Weizmann Institute of Science in Rehovot, Israel, and colleagues developed a device that looks like an oxygen delivery tube and detects various types of sniffing. Computer programs then translate the sniffs into typing and wheelchair movement. The team found that healthy subjects could use the device to type sentences and navigate a wheelchair by sniffing, as well (though not as fast) as they were able to using their hands.

With the wheelchair program, a quadriplegic patient moved the wheelchair as well as a healthy, nonparalyzed person could. When quadriplegic and locked-in patients tried the typing program, all used the typing program without much difficulty, increasing their speed and accuracy with each use, the researchers reported in the Proceedings of the National Academy of Sciences.

8. # Is Pharma Running Out of Brainy Ideas?

1. Greg Miller

Recent cutbacks raise concerns about the future of drug development for nervous system disorders.

When Emiliangelo Ratti found out late last year that GlaxoSmithKline (GSK) planned to pull the plug on drug discovery in some areas of neuroscience, including pain and depression, he knew he had to do something. As a senior vice president and head of the pharmaceutical company's center for drug discovery in neuroscience, Ratti oversaw work at two centers that were targeted for closure, one in Harlow, U.K., and one in Verona, Italy. Ratti scrambled to arrange a deal with an American contract research organization called Aptuit, which took over the Verona facility on 1 July and will provide research for hire for GSK and other companies. “I'm very proud of that because I've been able to secure the future of my 500 people,” Ratti says. At the Harlow facility, hundreds of employees have been laid off, while many others have been transferred within GSK, Ratti says.

In announcing the move to investors and analysts on 4 February, GSK Chief Executive Andrew Witty explained that pain, depression, and anxiety were areas where “we believe the probability of success is relatively low, [and] we think the cost of attaining success is disproportionately high.” Ceasing research in these areas would save GSK £250 million ($387 million) by 2012. A few weeks later, news came that AstraZeneca was closing research facilities in the United States and Europe and ceasing drug-discovery work in schizophrenia, bipolar disorder, depression, and anxiety. These cutbacks by two of the top players in drug development for disorders of the central nervous system (CNS) have raised concerns that the pharmaceutical industry is pulling out, or at least pulling back, in this area. In direct response to the cuts at GSK and AstraZeneca, the Institute of Medicine (IOM) Forum on Neuroscience and Nervous System Disorders organized a meeting in late June that brought together leaders from government, academia, and private foundations to take stock. (The forum's chair, Alan Leshner, is also the executive publisher of Science.) “The biggest problem isn't the announcements by GSK and AstraZeneca, it's when you look at the pipeline and see what companies are actually doing in psychiatric drug development,” says Thomas Insel, director of the National Institute of Mental Health. “There are very few new molecular entities, very few novel ideas, and almost nothing that gives any hope for a transformation in the treatment of mental illness.” That's worrying, Insel and others say, because the need for better treatments for neurological and psychiatric disorders is vast. Hundreds of millions of people are afflicted worldwide. Yet for some common disorders, like Alzheimer's disease, no truly effective treatments exist; for others, like depression, the existing drugs have limited efficacy and substantial side effects. ## What's in the pipeline? At first glance, the situation doesn't appear to be so dire. A report released 14 July by the Pharmaceutical Research and Manufacturers of America (PhRMA) touts a record-high 313 drugs in the pipeline for mental health disorders such as depression, anxiety, and addiction. Another report, commissioned by IOM for the June meeting and prepared by the Tufts Center for the Study of Drug Development, identified 1747 drugs in development for a much longer list of disorders, including degenerative diseases like multiple sclerosis and neurological conditions like epilepsy. Indeed, the Tufts report suggests that the pipeline has expanded rapidly for many conditions in recent years (see graph). But a closer look tells a different story, says Steven Hyman, a psychiatrist and former NIMH director who is now provost at Harvard University. Many of the drugs in clinical trials have long been approved and are now being tested for a new indication, Hyman says. Looking over the Tufts and PhRMA reports' lists of drugs in late-stage clinical trials for depression, he notes that both are loaded with antipsychotic drugs, including Risperdal (risperidone) and Seroquel (quetiapine), two of the first “atypical antipsychotics” approved by the Food and Drug Administration in the 1990s. “People with depression can have anxiety and agitation, and low doses of antipsychotics seem to improve those symptoms,” Hyman says. “But they don't necessarily have an independent effect on the core depressive symptoms, and they come with a real side-effect burden.” Other treatment candidates have limitations as well. Both lists include Corlux (mifepristone, better known as RU-486, the abortion drug). Even if it proves effective for depression, it couldn't be prescribed for women of reproductive age, Hyman notes. The Tufts list includes Agomelatine, a drug that boosts the effects of the hormone melatonin and blocks receptors for the neurotransmitter serotonin. Hyman says there's little compelling evidence that boosting melatonin has antidepressant effects, and he notes that the drug has had mixed results in European trials for depression. “This is hardly a rich pipeline,” Hyman says. “It suggests a sad dearth of ideas and involves lots of attempts at patent extensions and new indications for old drugs.” ## Risky business The reasons for the seeming lack of innovation are partly historical, says William Potter, who retired in January from Merck, where he was vice president for neuroscience. In the 1980s and '90s, drug companies realized that they could make billions of dollars a year off drugs that were slightly modified versions of already-approved medications, particularly the SSRI antidepressants like Prozac, Potter says: “The investment in truly innovative projects was not as deep as it might have been because you could make so much money from ‘me, too,’ drugs.” The current climate for innovation may be even worse. Companies will soon lose billions of dollars in revenue as patents expire on dozens of blockbuster drugs (see figure). Meanwhile, the costs of research and development are rising. “Most companies don't see where they're going to be getting the cash flow, so they're having to be more conservative,” Potter says. “You can't just ask companies to throw money at something that might not pay off.” The Tufts report suggests that pharma executives have good reason to see investments in CNS drug development as riskier than investments in other areas. CNS drugs cost more and take longer to bring to market than other types of drugs (see figure, right). And only 8% of CNS drugs that make it to clinical trials end up being approved, about half the average success rate across all therapeutic areas. Moreover, when CNS drugs fail, they tend to do so in late-stage clinical trials, after a significant investment has been made, says Kenneth Kaitin, director of the Tufts center. Adding to those troubles, the animal models, particularly for psychiatric disorders, are far from perfect at predicting which compounds will be effective in humans, and the clinical trials are often more complicated for CNS disorders, says Ratti. These disorders tend to be complex and intermittent, and their symptoms often defy objective measurement. “All these things together are making discovery and development in neuroscience significantly risky,” he says. Many companies see areas like oncology and autoimmune disorders as safer bets, says Steven Paul, who stepped down in February as executive vice president for science and technology for Eli Lilly and Co. “Their perception is that the science is a little richer and the odds are less daunting in some of these other areas,” he says. Even within neuroscience, Paul says, some companies may see psychiatric drugs as a bigger gamble than drugs for neurological conditions. Paul, who is a psychiatrist, says he doesn't necessarily agree with that assessment: “I personally believe there are compelling pathways and new targets to pursue.” ## Seeking a new model Given the economic challenges, experts inside and outside pharma say the old model of drug development, in which companies assume all the risks and costs of searching for new drugs and shepherding them from test tube to clinic, is no longer viable. “Traditionally, they would have significant internal research groups that would be as good as anyone at doing some of the basic research that might lead to new targets,” says Adrian Ivinson, who directs the Harvard NeuroDiscovery Center. “We're seeing less and less of that.” Many companies are trying to reduce costs by outsourcing R&D. AstraZeneca and GSK both set up R&D centers in Shanghai, China, in 2007, for example. The GSK center focuses on developing treatments for multiple sclerosis, Parkinson's, and Alzheimer's diseases; its work on those disorders will continue despite the recent cutbacks in other areas of neuroscience. Another outsourcing strategy involves contracting with biotech companies and academic researchers to do some of the early drug-discovery work that was previously done in-house. “They are sending scouts out into the community to talk to groups like ours and many others to identify projects with potential for drug discovery,” Ivinson says. In a typical arrangement, a company funds a research project in exchange for the right to license any resulting compounds that show therapeutic potential. At the IOM meeting, there was much discussion about public-private partnerships. Insel notes that the U.S. National Institutes of Health (NIH) already has a drug-discovery effort that might serve as a model, the Therapeutics for Rare and Neglected Diseases project launched last year. A measure in the new U.S. health-care legislation could expand NIH's role: It authorizes up to$500 million a year for a “Cures Acceleration Network” aimed at speeding drug development (Science, 26 March, p. 1562). “It's a strong message from Congress that they would like to see NIH more involved in drug discovery and drug development,” says Insel.

Another possibility might be a shared repository of compounds. “Pharma has thousands and thousands of compounds that are leads they've decided not to follow,” Insel says. “Would it make sense to put those into a resource that other people could begin to mine?” In May, GSK made public a library of potential malaria drugs, but neurology and psychiatry are far more lucrative markets. An open-access library of compounds would involve a host of challenges, not the least of which are questions about intellectual property, Insel says. “These are big and thorny issues, but we have got to grapple with them so that 10 years from now we're not looking at the same list of compounds that we know don't work well enough and hoping that if we just give them to a different group of patients we'll get a better outcome.”

9. Ecology

# Europe Tries to Save Its Eels

1. Gretchen Vogel

New national efforts to save the European eel face political and scientific challenges.

BERLIN—On a sunny morning this spring, German agriculture minister Ilse Aigner waded into the controversy over how to save the European eel. With several Berlin fishers, she helped release 65,000 young eels, called elvers, into the Havel River, which flows through Germany's capital city. Sporting rubber boots and a smile, Aigner gamely emptied her net of writhing fish into the cloudy river water. She then posed with the fishers, holding up a T-shirt that read “Eel Rescuer” and “Thank you for stocking eels!”

But whether Aigner's actions helped the long-term prospects for Anguilla anguilla, better known as the European eel, is hotly debated. Stocking eels in the Berlin river “is useless,” says Heike Vesper, fisheries officer for the conservation organization WWF Germany in Hamburg. “It's essentially a waste.”

Although the eels Aigner released have a good chance of thriving in the Havel, it is an open question whether any of them will reproduce. To contribute to the next generation, European eels must ultimately migrate back to the species' presumed spawning grounds in the Atlantic Ocean. From Berlin, that is a journey of more than 5000 kilometers, through dammed rivers, past deadly power plant turbines, and around fishing nets. Stocking young eels on the inland side of such barriers gives them only the slimmest chance of becoming a parent, say many conservationists and scientists.

No one disputes that eel rescuers are badly needed. Once abundant in European waters, the numbers of European eels have fallen drastically since the 1970s, by as much as 99% according to some estimates (Science, 10 October 2003, p. 221). Besides dams, turbines, and fishing, a number of factors, including a recently introduced parasite, chemical pollutants, and possible changes in ocean currents, are all suspected of playing a role in the eel's decline.

Sorting out the impact of each is difficult, because much about the eel's complex life cycle remains a mystery to scientists. “We are looking for life on Mars, but we still don't know where [European] eels spawn,” says fish physiologist Arjan Palstra of the University of Barcelona in Spain. The recent population crash, however, has stimulated innovative research on eels, from complex tracking experiments to simulations of their marathon migration in the lab, as scientists try to understand why the eel is disappearing—and to come up with ways that politicians and others might be able to make a difference.

Politicians are at least paying attention. In 2007, the European Union issued a regulation requiring each of its member countries to develop an “eel management plan.” The measure calls for limits on eel mortality based on a theoretical estimate of eel survival called the pristine rate: the number of eels that would have migrated back out to the open ocean in the absence of human impacts. The plans must ensure that at least 40% of this amount really do make the trip.

Nationwide conservation efforts based on the regulation are finally coming into effect; plans for two of the largest countries, Germany and the United Kingdom, were approved in April. But eel experts worry that many of the plans don't go far enough and fail to address some of the eels' most crucial problems. “From the perspective of the eel population, there's room for improvement,” says Kim Aarestrup of the National Institute of Aquatic Resources in Silkeborg, Denmark.

## Elusive eels

European eels apparently start and, if they survive long enough, end their lives in the deep, clear, calm, and featureless waters of the Atlantic Ocean's Sargasso Sea, a region bounded by major ocean currents and roughly the size of the continental United States. However, no one has seen a spawning European eel or its eggs in the wild.

The main evidence pointing to the Sargasso as the eels' spawning ground is that the smallest eel larvae have been spotted there. (One early observation was described in Nature in 1923, and that report is still regularly cited, notes David Righton of the Centre for Environment, Fisheries & Aquaculture Science in Lowestoft, U.K.) These larvae seem to grow from approximately 5 mm in the Sargasso to roughly 5 cm as they drift toward the European coast with the Gulf Stream. Scientists don't know whether larvae direct their course by actively swimming, and what they ate was a mystery too until recently—a study of larvae stomach contents reported in June indicate the eels consume zooplankton. By the time the animals get close to the coast, they have grown into 7-cm-long transparent eels.

These so-called glass eels enter estuaries from Morocco to Iceland. Unlike salmon, which return to their natal river or stream to spawn, eels seem to let ocean currents steer them by chance to their freshwater habitat. Some eels end up in Ireland, whereas others reach Greece; no genetic differences among these populations have been found.

The glass eels swim upstream and, once they enter fresh water, turn gray, becoming so-called elvers. Elvers gradually change color again, becoming yellow eels. The animals spend most of their lives in this form, living in rivers, lakes, and swamps and storing up energy, primarily as fat, for their final journey back to the Sargasso. Males spend roughly a decade as yellow eels, and females spend significantly longer—perhaps because egg production requires more energy than making sperm. Eventually, reacting to a trigger that scientists have yet to identify, they begin the long trip back to their birthplace to spawn.

To prepare for the journey, yellow eels undergo a radical transformation. They toughen their skin, alter their gills, and develop a black back and silver underside that helps them hide from open-ocean predators. The animals, called silver eels, rely entirely on their fat stores to make the long swim back to the Sargasso, ingesting only water on the way.

An average silver eel is more than 20% fat, making them a favorite target of fishers across northern Europe, where smoked, fried, and jellied eel are traditional favorites. (Yellow eels are also eaten in significant numbers.) In southern Europe, the fishers concentrate on the younger glass eels found in coastal waters. Some are shipped to Asia, where the spaghetti-thin eels are eaten as a delicacy or farmed for later harvesting as yellow eels. Others are sent to European aquaculture farms, where they are later sold to restocking programs—like the one Aigner highlighted in Berlin—or directly to fish farms where they are raised for European consumption. It is the number of glass eels that has fallen most dramatically over the past 30 years, arriving at less than 10% of the rate recorded in the 1970s. Once a cheap food, a kilo of glass eels now goes for about €500 ($640). ## Planning a comeback Conservationists worry that both silver and glass eel fisheries are unsustainable, given the dearth of incoming glass eels. But in political circles, there is some rivalry between the northern and southern fisheries. The tension—and the relative power of fishing interests—is reflected in the countries' different management plans as well. Ireland has taken the strictest approach, closing the commercial fishery at least through 2012. But some eel fishers can still make a living: Ireland pays them to “catch and carry” live eels over barriers, both on their inward and outward migrations. The Netherlands and the United Kingdom have shortened their eel-fishing seasons, banning fishers from catching glass eels especially in the last stages of their upstream migration and prohibiting fishing for silver eels during their autumn downstream migration. Germany's plan has been especially disappointing to conservationists, says WWF's Vesper. It relies primarily on stocking: transferring young eels from coastal regions—often from southern Europe—to habitats upriver. The strategy is based on fishing interests, not the eels' survival, Vesper charges. German stocking programs have been going on at least since the early 20th century and are traditionally paid for by fishing organizations. So as not to discourage those efforts, the German plan puts few new restrictions on eel fishing. Instead, it calls for the temporary shutdown of dam turbines and for an open season on another protected species: cormorants. Fishers blame the birds for eating a significant portion of the available eels. But some worry that stocking programs, although increasing the number of eels in Germany, may be detrimental to the species as a whole. No one knows exactly how successful stocked eels are at making their way back out to sea. Transporting young eels across the continent may disrupt their navigational imprinting and leave them unable to find their way back to the Sargasso, Palstra says. Reinhold Hanel of the Institute of Fisheries Ecology in Hamburg, Germany, and his colleagues are addressing that issue by analyzing eels' otoliths, structures in the inner ear that help animals maintain balance. Scientists can study the composition of the otoliths to determine where an eel has lived. From such information, Hanel's team is trying to determine what fraction of mature eels entering the Baltic Sea also spent their early years there. The project “should help us determine whether stocking is worthwhile or counterproductive,” Hanel says. Farther south, France has a plan to reduce fishing mortality by 30% and improve eel habitat and migration routes, but it does not ban the (very lucrative) export of eels to Asia, a step recommended by the International Council for the Exploration of the Sea. E.U. officials have not yet given the green light to plans submitted by Greece, Portugal, Spain, and Italy. Because the Mediterranean countries have little historical data available to estimate how many adult eels should be migrating out each year, regulators have been arguing over how many outgoing eels will fulfill the mandate. ## A fish odyssey A major weakness of all the European plans, say Palstra and others, is that they do not take into account the variable quality of silver eels leaving each country on their reproductive migration. “[Potential] spawners are leaving everywhere from Morocco to Iceland. We don't know which of those really contribute to the next generation,” says Russell Poole of Ireland's Marine Institute in Furnace. The target rate of outgoing eels may need to be adjusted upward if the viable breeding fraction is low. To try to help answer that question, Righton, Aarestrup, Poole, and colleagues from seven countries across Europe are cooperating on a 4-year, €4 million ($5.2 million) project called the “eeliad,” which is tagging migrating silver eels as they start their journey. The eels carry 12-centimeter-long satellite tags to record water depth, temperature, sunrise, and sunset. The tags release at a preset time, rise to the surface, and transmit their stored data. No tags have yet made it all the way to the Sargasso, but an earlier tagging project found that migrating eels spend the night in shallow, warmer water and dive during the day to depths of up to 1000 meters. Such dives might help the animals regulate their temperature, allowing females to postpone maturation of their eggs until they reach the Sargasso Sea (Science, 25 September 2009, p. 1660).

Difficulty moving up and down in the water column may be one reason the European eel is in trouble. Eels use an organ called the swim bladder to regulate buoyancy. This organ is attacked by a nematode parasite, called Anguillicola crassus, that started appearing in European waters in the 1980s, presumably imported from Asia along with Asian eels used in aquaculture. The parasite has since spread across Europe, given a lift by stocking efforts. Whereas Asian eels have some natural resistance to the parasite, usually hosting just five or six per animal, infected European eels frequently have dozens. To better ascertain how the parasites affect this migration, eeliad researchers last year tagged eels known to be infected, Righton says.

The parasite problem can also be explored in the laboratory thanks to 22 eel “swim tunnels,” a complex of tubes filled with 7000 liters of water that allows the fish to swim thousands of kilometers as if they were returning to the ocean. Using this device, Palstra, his Ph.D. supervisor, Guido van den Thillart of Leiden University in the Netherlands, and their colleagues found in 2007 that eels infected with the parasites swim 20% slower while using 20% more energy than uninfected eels—perhaps using up their fat reserves before they reach the Sargasso or causing them to have little left over to devote to egg production once they arrive.

The international eel trade has also spread a suite of viruses that render eels less fit for their epic journey. Van den Thillart and Vincent van Ginneken, also of Leiden University, have found that eels infected with so-called Eel Virus European X developed hemorrhages and anemia during their simulated migration in the swim tubes and died after swimming roughly 1000 kilometers.

So it is possible that even if the management plans allow significant numbers of silver eels to escape back into the Atlantic, relatively few are able to complete their biological mission. Palstra suspects that in the long term, Europe may need to raise wild-caught young eels in clean habitats—free of viruses, parasites, and chemicals such as PCBs—and ensure that they are able to leave for the open ocean.

Breeding eels in captivity would be one way to reduce fishing pressure on the wild population, but for now, this is still just a research goal. Scientists have had some success: They can use hormone treatments to prompt European eels to produce sperm and eggs in the lab. The eggs fertilize and hatch, but the larvae die after a few days, in part because no one has yet figured out what to feed them. Japanese researchers announced in April that they had produced offspring from Asian eels that had been bred and raised in captivity, for the first time completing an eel's life cycle in captivity. They apparently feed their larvae a slurry of shark egg powder, Palstra says, but their success rate is still very low. He and colleagues from 15 institutes from Norway to Tunisia are part of a new E.U.-funded project, PRO-EEL, that aims to replicate the Japanese success with European eels.

## Blame it on nature?

A few scientists question whether humans are really at fault for the crash of the European eel. The decline could be part of a natural cycle, says Brian Knights, a retired eel biologist from the University of Westminster. Perhaps changes in the Gulf Stream are driving the larvae off course or changes in water temperature or food availability are affecting the animals' survival. “It might be that the major cause of change is beyond our control,” he says.

Most eel experts, however, don't hesitate to point the finger at people's appetite for the fish. “If you look at the exploitation of eels, habitat loss, and disease, … it's hard to argue that human impact has nothing to do with it,” counters Aarestrup. “But because we know so little, it's hard to make the argument” that particular interventions will make a difference, he admits.

And eel watchers agree that there will not be any quick payoffs from Europe's conservation plans. In many regions, it can take female eels 20 years to mature into migration-ready silver eels. “You're looking at a very slow recovery process for the population,” says Poole. “You're not going to see anything obvious in the next 20 to 50 years.”

At the same time, it is unlikely that the European eel will go completely extinct. “Anyone who talks about extinction is alarmist,” Poole says. “You're talking about fish that spawn multiple millions of eggs.” The fisheries, however, may face real danger. “You can't continue to exploit a stock where current recruitment is less than 10% of previous levels,” Poole says.

The political decisions are not easy ones, Aarestrup admits. “We know less about eels than we know about climate changes,” he says. “But the question is, can we afford to do nothing?”

10. Exploration

# Making Smarter, Savvier Robots

1. Sam Kean

What machines of the future really need to learn, say experts who plan to have them explore the far reaches of the solar system, is more independent behavior.

On 18 July 2009, the Mars rover Opportunity was scooting toward a distant martian crater when it spied an anomaly amid the ripples of red soil: a bruise-colored rock the size of a watermelon. It looked like a meteorite— potential evidence that the ancient atmosphere of Mars, like today's, was thin enough for such rocks to pass through without exploding.

The strange rock was exactly the kind of thing NASA sent Opportunity to find. But because Mars and Earth are millions of kilometers apart and rotate out of sync, NASA scientists didn't see it until Opportunity had driven 200 meters beyond it. They hit reverse but had to wait three full days for Opportunity to backtrack to the spot.

The researchers got their meteorite. But the near miss—and the frustrating delay—underscored a defect of current exploration technology: Basically, robots are pretty dumb. Now scientists across the world are striving to change that by developing intelligent robots that can circumvent danger and spot enticing features on their own.

Hundreds of scientists, mostly at NASA and at universities, are working on improving robot explorers. But only a few dozen specialize in developing robots with true, high-level independence. The main NASA lab, at the Jet Propulsion Laboratory (JPL) in Pasadena, California, has a dozen people and a budget of about \$4 million—a lower figure than in the past. But scientists there see promising signs. For one, NASA chief technologist Robert Braun has begun a new, general Space Technology Program that lists “machine intelligence” as one thrust.

“There are compelling reasons to send humans into deep space,” says Steve Chien, who develops autonomous space systems for JPL. “A smart scientist can do much better experiments. But it's very expensive. By making the spacecraft much smarter, we can reduce the gap between human exploration and robotic exploration.”

## Where to go

Robots with an IQ boost will be essential for fully exploring some locations in the solar system—including hostile spots. On Venus, for example, 450°C surface temperatures and pressures comparable to those a kilometer deep in the ocean will destroy the onboard computers of any lander within 5 hours, tops. To get anything done, the lander will need to perform experiments, such as sampling soil, without human input.

Rendezvous missions with comets or asteroids and landings on distant moons would also benefit from more autonomous robots, researchers say. On Saturn's moon Titan, radio waves carrying scientists' instructions take 90 minutes to arrive from Earth. Yet a probe flying through Titan's atmosphere would have to negotiate hazards in real time, notes Wolfgang Fink, a computer scientist working at the California Institute of Technology and the University of Arizona. “If it's about to fly into a mountain range, it can't say, ‘I'm flying into a mountain range. Please advise,’ and wait 1½ hours.”

Scientists also hope that greater intelligence will make robots more efficient, improving their “energy storage, memory, computational throughput, communication downlink bandwidth, and heating and cooling capability,” says Larry Matthies, a computer scientist at JPL. Opportunity (and its companion on Mars, Spirit) travel at such pokey paces—28 kilometers total in 6 years—partly because they rely on humans to spot dangerously loose sand or steep slopes. A smarter robot could zip around obstacles by itself and travel up to 10 times as far each day, Matthies estimates. And the more work the rover can do alone, the more time it will have to collect good samples.

## Recipes for “eureka”

In December and January, NASA took the first steps toward making a spacecraft autonomous when it uploaded four pieces of software to Opportunity. Tara Estlin, a senior engineer at JPL, explains that, with the new software, “scientists can give us a single property or combination of properties—the largest rock you can find, or the darkest rock,” and Opportunity will zero in on them. In March, the software passed its first test by discovering, all on its own, an angular, football-size rock—ejected from a nearby impact crater—in a field of rounder boulders. (Paradoxically, though, Estlin's team still has to tell the rover a day in advance when to be autonomous and when not to.)

Earth-based systems have already demonstrated significant independence, within limits. Chien works on the Earth Observing Sensorweb, a group of half a dozen NASA satellites that monitor Earth's atmosphere. Some scan large sections of Earth's surface and pick out a flood or a volcanic plume from space. They beam the data to ground-based computers, which in turn direct higher-resolution satellites to focus on the event—all without human input. Chien hopes to expand the work to other planets. But the instruments can spot only a short list of predefined events; they cannot find anything interesting or new on their own. Asked whether the system could shift its attention on its own between the two most notable geological events of the past few months, the Eyjafjallajökull volcano eruption in March and the BP oil slick in the Gulf of Mexico in April, Chien groans: “I wish, I wish.”

To solve problems of data filtering and interpretation, some researchers are working to cultivate a robot's taste for the unusual. Sometimes scientists want to study the most representative feature around, but more often they are intrigued by anomalies. “If the whole desert is smooth and one area is rough, that's interesting,” says Chien. “If the whole desert is rough and one area is smooth, that's interesting. If you really don't know about the environment, you have to fall back on something like outliers.”

Patrick McGuire, a computer scientist and geologist at the University of Chicago in Illinois, has developed a simple setup that can detect novel features in a landscape. A netbook laptop hooked up to a cell phone with a camera snaps a picture and compares its colors, textures, and shapes with other pictures in its memory. The computer then compresses the image with an algorithm. If the compression process is very similar to that of an earlier image, the computer concludes that the new image doesn't contain much novel information and throws it out.

McGuire has tested this system at rock outcrops in Utah and Spain that resemble the barren landscape a probe might encounter on a distant planet. He reported late last year that the software performed equally well in both locations. In one case, the software immediately recognized a patch of lichen as novel—and then, with the next picture, threw out an image of lichen on another rock as too similar to bother remembering, demonstrating that it is a quick study.

## Curious future

Some scientists, including Fink, say better programming alone won't turn robots into independent explorers. “In planetary exploration, you're in for surprises,” Fink argues, “and you will not always have a rule” on how to proceed. He dreams of robots that can experiment with their own “neural networks”— their internal architectures for taking inputs, processing information, and producing outputs—and can, like humans, form their own rules for exploring.

McGuire says certain architectures have advantages in different applications. With a so-called Hopfield neural network, for example, a computer can recognize an entire picture stored in memory after seeing only a fraction. Many robots come equipped with multiple lenses and cameras that take pictures on different scales, so the capability to tag small snippets as familiar would help make the robot more efficient in selecting which scenes to shoot or not shoot.

Even more ambitiously, Fink is developing systems to give robots freedom to change their logical architecture—essentially to “rewire” their brains. A robot might make a rule more complicated or simpler by adding or cutting steps, or combining the binary code of two rules and trying out their “offspring.” If the new rules worked well, it adds them to its problem-solving repertoire.

Fink published a paper last year on a self-configuring neural network to sort odd numbers from even numbers. Working with numbers 24 bits long—in the tens of millions in decimal notation—the network hit upon the solution (look at the final digit) with no guidance. And by focusing on one bit, the computer freed 23 other bits for different tasks. In other contexts, such self-configuring networks have helped scientists design circuits and new drugs.

Ultimately, Fink says, he hopes to instill something like curiosity in robots. That kind of programming would go far beyond algorithms his team has developed to help robots calculate the best angle to stretch out an arm to grasp an object or scoop soil. “We're after the intent to deploy the arm. How does the spacecraft know where it wants to dig? This is of interest to me.”

## The first test?

Smart, curious robot explorers wouldn't have to work alone. Fink envisions a multitier scheme of robots with satellites, blimps or balloons, and platoons of ground rovers. An intelligent satellite would direct the blimps to canvass certain areas. The blimps, in turn, would direct surface rovers to scout hydrothermal vents or rappel down cliff faces with a cable. Based on feedback between each tier, the satellite would decide which sites to concentrate on and how best to deploy the other machines. It would judge when to risk sending rovers into dangerous areas like active volcanoes, and when to stop collecting data, Fink says. “A spacecraft could even leave a place and tell you, ‘There's nothing interesting here. I'll go somewhere else and I'll tell you when I get there.’” Fink and his team have started building a test site in Arizona with rovers, boats, and blimps for field experiments with rudimentary versions of such robotic expeditions.

Chien imagines a different sort of teamwork: human explorers with fast-learning robot assistants. A group led by David Akin at the University of Maryland, College Park, is testing a golf cart–size three-wheeled rover, named Raven, to help astronauts explore planets. On tricky terrain such as loose soil or slopes, Akin says, the astronaut can simply say, “Follow my path,” and the robot will.

President Barack Obama's stated goals of sending humans to an asteroid in the 2020s and to Mars in the 2030s could help foster such partnerships. Chien says human-and-robot teams could do a better job together than either could alone. Humans would make plans and be in charge, while robots slogged through the important but routine technician work. “It's the classic apprentice thing,” Chien says. “You want the biggest brainpower worrying about the biggest problems.”